High-Dose Heparin Decreases Nitric Oxide Production by Cultured Bovine Endothelial Cells
Background Abrupt cessation of heparin therapy can lead to a recrudescence of thrombosis and acute ischemia. Endothelial NO is an important endogenous inhibitor of platelet-mediated thrombosis, yet biochemical studies examining the effect of heparin on NO production by the endothelium have heretofore been lacking.
Methods and Results In an attempt to address the effect of heparin on endothelial cell production of NO, confluent bovine aortic endothelial cells (BAECs) on microcarrier beads were incubated in the presence or absence of heparin. Results indicate that BAECs incubated with heparin were less able to inhibit platelet aggregation than control cells (P<.005 by ANOVA) and that this effect correlated with a decrease in NO production (36% decrease for heparin compared with control, P<.05). Dextran sulfate evoked the same response (67% decrease, P<.0001 compared with control), suggesting that the decrease in NO after heparin treatment is secondary to its negative charge rather than to a specific polysaccharide sequence. The decrease in NO production by heparin was accompanied by a 72% decrease in steady-state Nos 3 mRNA as well as a 49% decrease in immunodetectable endothelial NO synthase (eNOS) protein.
Conclusions These data show that high-dose heparin at concentrations achieved in some acute cardiovascular settings increases in vitro platelet aggregation in media conditioned by endothelial cells by decreasing endothelial NO production through a mechanism that involves a decrease in steady-state Nos 3 mRNA and eNOS protein. These observations suggest a possible mechanism by which to explain in part the prothrombotic effects of heparin.
Following the initial report by Howell,1 Best et al2 showed in 1938 that heparin, which had been isolated from dog liver, was capable of preventing the formation of a “white” thrombus. Today, heparin is the primary anticoagulant used in the setting of acute coronary syndromes and in percutaneous transluminal coronary angioplasty and most cardiovascular procedures, including cardiopulmonary bypass surgery. This polydispersed polysaccharide exerts its effects on the coagulation system primarily by catalyzing the inhibition of procoagulants, such as thrombin, by the serine protease inhibitors (serpins) antithrombin III and heparin cofactor II.3 4 5 The inactivation of thrombin by these serpins leads to both an extended activated clotting time and altered platelet function as well as a heparin dose–dependent decrease in arterial platelet deposition after balloon injury.6 7
Recent studies have also shown that heparin affects the fibrinolytic system as well as platelet function in vivo.8 9 10 Studies by Fareed and colleagues8 demonstrated that heparin administered before the initiation of cardiopulmonary bypass resulted in an increase in plasmin activity. Using immunoblotting studies of plasma from baboons administered increasing doses of heparin, we recently demonstrated that heparin leads to a dose-dependent increase in plasmin light chain in vivo as well as the appearance of a specific fibrin degradation product, fragment E.10
In contrast to the role of heparin as an anticoagulant and profibrinolytic agent, this glycosaminoglycan has also been shown to promote a prothrombotic state under certain conditions.11 12 13 Landolfi and colleagues11 demonstrated in vivo that platelets from patients who had received unfractionated heparin secrete increased levels of thromboxane A2, an activated platelet product. Other studies have shown that heparin infusion ex vivo can induce platelet aggregation independent of antibody-mediated heparin-associated thrombocytopenia and thrombosis and that this mechanism may be important in patients with acute myocardial infarction who receive thrombolytic therapy.12 Chen and colleagues13 also demonstrated that low-molecular-weight heparin induces platelet aggregation, whereas hirudin, a direct thrombin inhibitor, fails to induce platelet aggregation. Abrupt cessation of heparin infusion in patients with unstable angina can lead to an acute recrudescence of thrombosis and coronary ischemia, an effect that can be prevented by the concomitant use of aspirin.14
EDRF, first described by Furchgott and Zawadzki15 and now believed to be authentic NO16 or an adduct of NO,17 plays an integral role in modulating platelet adhesion, activation, and aggregation.18 The endothelium regulates platelet-platelet and platelet–vessel wall interactions by altering its production of NO and thereby preserves the antithrombotic properties of the normal vasculature. Although a number of physiological studies have attempted to examine the effects of both heparin and the heparin/protamine complex on vasoreactivity,19 20 21 studies directly examining the effect of heparin on the production of NO by endothelial cells have been lacking. We therefore hypothesized that heparin impairs endothelial production of NO, which in turn promotes platelet activation and thrombosis.
Chemicals and Buffers
ADP, acetylsalicylic acid, bradykinin triacetate, l-arginine, Na2EDTA, sodium nitrite, HEPES, actinomycin D, Chelex anion exchange resin, reduced glutathione, and protamine sulfate (grade X) were purchased from Sigma Chemical Co. Heparin (porcine intestinal) was purchased from Elkins-Sinn, Inc, with a weight-average molecular size of 15 000 kD (5 U/mL ≈1.85 μmol/L heparin). Recombinant hirudin was obtained from CIBA-Geigy. Dextran sulfate (sodium salt) and microcarrier beads (Cytodex 3) were purchased from Pharmacia LKB. L-NMMA was obtained from Calbiochem. cGMP was measured by an ELISA from Cayman Chemical Co. HBSS without calcium or magnesium, NCS, antibiotics, trypsin, and media were all purchased from Gibco BRL. PBS (pH 7.4) consisted of 10 mmol/L sodium phosphate and 150 mmol/L NaCl, which was treated with Chelex before use. Total protein concentration was determined with a BCA (bicinchoninic acid) Protein Assay Reagent Kit obtained from Pierce Chemical Co.
BAECs were isolated as previously described.22 BAECs were maintained in DMEM/F12 containing 20% NCS and antibiotics (100 U/mL penicillin G sodium and 100 μg/mL streptomycin sulfate). Cultures were maintained in a humidified incubator at 37°C with a 5% CO2 atmosphere. Cells at passages 3 through 12 were subcultured after treatment with 0.05% trypsin and 0.53 mmol/L EDTA. BAECs were identified by their maintenance of density-dependent growth after serial passage, by their typical cobblestone configuration when viewed by light microscopy, and by a positive indirect immunofluorescence test for von Willebrand factor.23 24 25 Viability of confluent cell monolayers was determined after exposure to heparin (known to be angiostatic26 ) by examination of the cells at the end of each treatment by light microscopy for vacuolation and other signs of cell death. All experiments were performed in phenol red–free DMEM/F12 with 20% NCS and 1 mmol/L l-arginine, the substrate for NO synthase.
Preparation of ECMBs
BAECs were grown to confluence, then coated on a microcarrier bead system (Cytodex 3) as previously described.27 In these experiments, ECMBs were treated for 4 hours with either 0 or 5 U/mL heparin. Before use, ECMBs were exposed to 30 μmol/L acetylsalicylic acid at 37°C for 45 minutes, washed three times, and resuspended in PBS.28 29 A packed volume of 500 μL ECMBs was calculated to deliver 3182±223 beads/droplet.
Preparation of PRP and GFPs
Briefly, venous blood was drawn into 13 mmol/L sodium citrate from volunteers who had not ingested acetylsalicylic acid for at least 7 days, as previously described.28 PRP was prepared by centrifugation, with platelet counts determined with a Coulter counter (model ZM, Coulter Electronics). The platelet count was adjusted to 2.0×105 platelets/μL by the addition of platelet-poor plasma.
GFPs were prepared from blood collected from volunteers in citrate-phosphate-dextrose (1:10). After centrifugation at 120g for 10 minutes, GFPs were prepared by passing PRP over a Sepharose 2B column that had been washed earlier with 2 column volumes of deionized water and 20 column volumes of a HEPES buffer consisting of (in mmol/L) HEPES 5.8 (pH 7.4), NaCl 140, KCl 6.11, MgSO4 2.5, Na2SO4 2.44, and dextrose 5.64, and 5 μmol/L BSA. Platelet counts were adjusted to 200 000 platelets/μL with HEPES buffer.
NO Generation From Cultured Endothelial Cells as Determined by Platelet Aggregation Studies
On the basis of earlier calculations,28 29 one, two, or three droplets of ECMBs (4.45×106, 8.9×106, or 13.35×106 endothelial cells, respectively) were added to PRP and stirred in a cuvette at 500 rpm for 3 minutes within a four-chambered aggregometer (BioData Corp) at 37°C. ECMBs were allowed to settle, and 300 μL PRP free of ECMBs was removed for immediate aggregation assay. Platelet aggregation was induced with 5 μmol/L ADP and monitored by a standard nephelometric technique.28 Aggregation at 37°C was quantified by measurement of the extent of light transmittance and is expressed in percent.
Heparin Treatment of Cultured Endothelial Cells
BAECs were plated onto 20×100-mm tissue culture dishes (Falcon 3003) and allowed to reach confluence over a period of 5 to 7 days. Phenol red–free media with 20% NCS and 1 mmol/L l-arginine were replaced to initiate experiments. Initially, confluent BAECs on 20×100-mm tissue culture plates were treated with increasing concentrations of heparin (0, 0.5, or 5 U/mL). Media were collected at 4 hours and assayed for NO as described below.
Measurement of NO in Media
The production of NO (S-nitrosothiols and free NO) was measured by photolysis-chemiluminescence.30 31 32 Briefly, the system detects NO using ultraviolet light to cleave the S-NO bond homolytically; free NO is then detected in a chemiluminescence spectrometer by reaction with ozone. SNO-Glu was used as the reference standard. All measured NO is expressed as mean NO (nmol/g cell protein)±SEM.
Effect of Heparin on Detection of SNO-Glu in Media
Because of a concern that heparin might interfere with the measurement of NO as determined by photolysis-chemiluminescence, we examined the effects of heparin on an exogenous NO donor, SNO-Glu, in media. Briefly, 1 μmol/L reduced glutathione in 1 mol/L HCl was mixed (1:1 vol/vol) with 1 μmol/L sodium nitrite. The resulting SNO-Glu was added to increasing concentrations of heparin in medium, and NO concentrations were determined as described above.
Effect of Dextran Sulfate, Protamine, or Hirudin on NO Production by Endothelial Cells
Having demonstrated a significant decrease in the production of NO by BAECs after exposure to heparin at pharmacological doses, we attempted to determine whether the decrease in NO production was specific for a heparin saccharide sequence or the consequence of a nonspecific polyanionic effect. Dextran sulfate at concentrations (1.85 μmol/L) equimolar to that of 5 U/mL heparin was added to phenol red–free media containing 20% NCS and 1 mmol/L l-arginine. After 4 hours, media were collected and assayed for NO. The cells were washed with HBSS and treated with 1N NaOH, and total protein concentrations were determined.
In separate experiments, 1.85 μmol/L dextran sulfate and 1.85 μmol/L protamine sulfate, a polycation, were combined (1:1 vol:vol) and added to media. Confluent BAECs were then treated for 4 hours with either no exogenous polyion, the dextran/protamine solution, or protamine alone. Media were collected and assayed for NO, and total protein concentrations were determined.
Additional experiments aimed at determining whether negative charge alone could alter NO production by BAECs were performed with recombinant hirudin, a polyanionic peptide. In these experiments, BAECs were treated with 1.85 μmol/L hirudin for 4 hours, and NO levels were determined.
Endothelial NO and Platelet cGMP
Confluent BAECs were treated with media containing 0 or 5 U/mL heparin for 4 hours. At the end of this incubation period, 3 mL GFPs were incubated with the cells. After 30 minutes, the GFPs were collected, washed with PBS, and acid-precipitated with cold 10% trichloroacetic acid. GFPs at 37°C for 30 minutes without exposure to endothelial cells served as a control. cGMP was measured with an ELISA.
Northern Analysis of eNOS (Nos 3) mRNA
Cell monolayers of ≈1.1×107 cells were grown to subconfluence. Total RNA was isolated from monolayers of BAECs treated for (1) 4 hours with media containing no exogenous heparin or (2) 4 hours with media containing 5 U/mL heparin. Total RNA was extracted by the Oncogene Science RNA purification system, which is based on a guanidinium thiocyanate–phenol-chloroform extraction procedure.
Steady-state endothelial NO synthase (Nos 3) mRNA was detected by Northern analysis using a full-length cDNA insert derived from a bovine clone kindly provided by Drs Thomas Michel and Santiago Lamas, Brigham and Women’s Hospital, Boston, Mass (accession No. M89952).33 Total RNA (10 or 20 μg) was loaded in each lane and electrophoresed on a denaturing 1.2% agarose gel containing 2.2 mmol/L formaldehyde. The gel was blotted onto a Nytron-N+ membrane by capillary action, hybridized with the [32P]-radiolabeled probes for Nos 3 and bovine β-actin (generous gift of Dr David R. Morris, University of Washington, Seattle), and then exposed on Kodak X-OMAT film for 24 or 48 hours at −70°C. For quantitative evaluation of Nos 3 and β-actin transcripts, phosphoimage analysis was performed with a PhosphoImager SF (Molecular Dynamics). The blots were repeated four times for each of the two experimental groups and the results averaged.
After demonstrating a significant decrease in steady-state Nos 3 mRNA in BAECs treated with heparin, we attempted to determine whether this effect was secondary to differences in transcription or message stability. To this end, BAECs were treated with 0 or 5 U/mL heparin for 4 hours followed by the addition of 5 μg/mL actinomycin D to the medium. Total mRNA was collected as described above at 0, 1, 2, and 4 hours, and Northern analysis was performed.
Western Blot Analysis of eNOS
Confluent BAECs treated with either control media or media containing 5 U/mL heparin for 4 hours were harvested by scraping. After centrifugation, the cells were frozen at −70°C overnight. Upon thawing, the cell pellet was homogenized in a buffer containing the following (in mmol/L): sucrose 0.32, HEPES 20, EDTA 0.5, and dithiothreitol 1; (in μmol/L): leupeptin 2.0, pepstatin A 1.0, and PMSF 1.0. Insoluble material was sedimented at 1000g for 10 minutes at 2°C, and a sample of each supernatant was used for determination of total protein concentration. The supernatants were then further denatured by boiling for 2 minutes before electrophoresis.
Each sample was loaded onto a 4% stacking/7.5% separating gel at equal protein concentrations and electrophoresed at constant current overnight at 4°C. One half of the gel, which had been loaded in duplicate, was stained with Coomassie blue to confirm equal protein loading, while the other half of the gel was transferred for immunoblotting. After transfer, the blot was placed in 5% milk protein blocking solution overnight to block nonspecific-binding sites. The blot was then exposed to a human monoclonal antibody to eNOS (1:2000, Transduction Laboratories). The ECL Western blot analysis system (Amersham Life System) was used for detection and included a peroxidase-labeled anti-mouse secondary antibody (1:1000). The immunoreactivity of eNOS was detected by changes in chemiluminescence. After transfer to an autoradiograph, the eNOS signal was quantified by densitometry.
Paired samples were compared by Student’s t test. Groups of data were compared across time and concentration by repeated-measures ANOVA when multiple experiments were performed, with a Dunn’s or Newman-Keuls post hoc test. All data are presented as the mean±SEM, with P<.05 considered to be statistically significant.
Heparin Treatment of Endothelial Cells Attenuates EDRF-Dependent Platelet Inhibition
The effect of heparin treatment (5 U/mL for 4 hours) on EDRF/NO production by BAECs was first studied by a bioassay of the inhibition of platelet aggregation. With cells that had been treated with control media, platelet aggregation was progressively inhibited with increasing numbers of endothelial cells (86%, 69%, and 33%* light transmittance for 4.5×106, 8.9×106, and 13.4×106 endothelial cells, respectively, *P<.05 by ANOVA compared with 86% and 69%). However, when BAECs were treated with media containing 5 U/mL heparin, the progressive decrease in platelet aggregation was blunted (97%, 92%, and 77% light transmittance, respectively) (Fig 1⇓). Compared with BAECs treated with control media, cells treated with heparin were significantly less able to suppress platelet aggregation (P<.005 compared with control). The inhibitory effects of BAECs on platelet aggregation were prevented by incubation with 100 μmol/L L-NMMA in the absence (95±5% transmittance) or presence (96±6% transmittance) of 5 U/mL heparin.
Heparin Decreases NO Production by BAECs in a Dose-Dependent Manner
After demonstrating that BAECs treated with heparin are less able to inhibit platelet aggregation, we treated BAECs with increasing concentrations of heparin in an attempt to examine the effects of heparin on NO production directly. BAECs were treated with media containing either 0, 0.5, or 5.0 U/mL heparin for 4 hours, and the results obtained are presented in Table 1⇓. The results demonstrate that heparin decreases BAEC production of NO (to 98% and 64%* of control, respectively, *P<.05 compared with control by repeated-measures ANOVA, n=3 in triplicate). In BAECs treated with 5 U/mL heparin, a concentration of heparin attained during cardiopulmonary bypass,31 NO was significantly decreased (81±21 nmol/g protein for 5 U/mL, P<.05 by Dunn’s method) compared with cells treated with control medium (123±16 nmol/g protein for control).
In other experiments, heparin (0.5, 1, 5, 10, and 100 U/mL) was added to medium containing 1 μmol/L SNO-Glu to determine whether the polyanion could interfere with the measurement of NO produced by BAECs. The results demonstrate that increasing concentrations of heparin do not significantly alter measured SNO-Glu in medium compared with control, as determined by chemiluminescence with each experiment performed in triplicate (data not shown).
Treatment of BAECs With Heparin Reduces NO-Stimulated Platelet cGMP Production
BAECs were incubated with GFPs for 30 minutes after treatment with 0 or 5 U/mL heparin for 4 hours, and cGMP content of the platelets was determined. Compared with control GFPs incubated at 37°C without BAECs, cGMP levels in GFPs exposed to BAECs after 0 U/mL heparin were increased by 150%. In contrast, cGMP levels in GFPs exposed to BAECs after 5 U/mL heparin sustained only a 56% increase compared with control GFPs. L-NMMA completely suppressed cGMP increases in both cases (Table 2⇓).
Dextran Sulfate Decreases NO Production by BAECs, and This Decrease Is Reversed by the Addition of Protamine
We next treated cells for 4 hours with a concentration (1.85 μmol/L) of dextran sulfate equimolar to that of heparin to examine the saccharide sequence specificity of the observed effect. Total NO was determined as described in “Methods.” Results, presented in Fig 2⇓, demonstrate that cells treated with dextran sulfate show a 67% decrease (P<.0001) in NO production compared with control.
After demonstrating a statistically significant decrease in NO production after treatment of BAECs with dextran sulfate, we next treated cells with 1.85 μmol/L dextran sulfate in the presence of an equimolar concentration of protamine sulfate, a polycation, to determine whether charge neutralization of the polyanion alters its effect on NO production. The results, presented in Fig 2⇑, show that BAECs treated with dextran sulfate, the charge of which had been at least partially neutralized by protamine, decreased NO production by only 21% compared with cells treated with control medium (109±15 nmol NO/g protein for control versus 86±13 nmol NO/g protein for dextran+protamine, P=NS, n=5 per group). BAECs treated with 1.85 nmol/L protamine demonstrated no significant change (93±5 nmol/g protein for protamine) compared with BAECs treated with control medium. Similarly, BAECs treated with 1.85 nmol/L hirudin, a polyanion polypeptide, failed to produce a significant decrease in NO (93% of control).
Heparin Decreases Steady-State Nos 3 mRNA in BAECs
The effect of heparin on steady-state Nos 3 mRNA was quantified by Northern analysis. Total cellular RNA was obtained after treatment of BAECs with either control medium or medium containing 5 U/mL heparin, each for 4 hours. Fig 3⇓ is a representative Northern blot of BAEC mRNA probed with a full-length Nos 3 cDNA probe and a bovine β-actin cDNA probe (4.5 and 2.0 kb, respectively). Results show that cells treated with 5 U/mL heparin have a 72% decrease in steady-state Nos 3 mRNA compared with control cells after normalization for β-actin.
In an attempt to determine whether the decrease in steady-state Nos 3 mRNA was secondary to an attenuation in transcription or a decrease in mRNA stability, cells were treated with actinomycin D after 0 or 5 U/mL heparin for 4 hours as described in “Methods.” Fig 4⇓ demonstrates no significant change in Nos 3 mRNA signal over time, as determined by phosphoimage analysis, between control and heparin-treated cells, suggesting that heparin is affecting Nos 3 mRNA transcription. No change in β-actin mRNA was observed in these experiments, given its long half-life (data not shown).
Decrease in eNOS Protein Levels After Heparin Exposure
To ascertain the significance of the decrease in steady-state Nos 3 mRNA after heparin exposure for eNOS protein expression, Western blot analysis was performed on cells treated with either control medium or medium containing 5 U/mL heparin. The results presented in Fig 5⇓ demonstrate that cells exposed to 5 U/mL heparin express 49% less eNOS protein than control cells, as determined by densitometric analysis. Importantly, in all of these experiments, there was no effect of heparin exposure on endothelial cell viability, as measured by trypan blue uptake or lactate dehydrogenase release, compared with control cells.
In the present study, BAECs grown on microcarrier beads in the presence of 5 U/mL heparin were less able to inhibit platelet aggregation than BAECs grown in the absence of heparin, suggesting that a decrease in NO production by BAECs after heparin exposure potentially has clinical ramifications. We demonstrated that endothelial cells exposed to concentrations of heparin (5 U/mL) attainable during percutaneous transluminal coronary angioplasty or cardiopulmonary bypass34 respond by decreasing their NO production over time. In addition, this decrease in NO production by BAECs after heparin treatment resulted in a decrease in platelet cGMP. After this observation, using Northern and Western analyses, we demonstrated that heparin evokes its effects on NO production in BAECs by decreasing steady-state Nos 3 mRNA levels and eNOS protein levels.
In an attempt to discover the mechanism by which heparin affects NO production by BAECs, we examined the effects of various charged polyionic species, namely dextran sulfate (polyanion) and protamine sulfate (polycation), on NO production by endothelial cells. A decrease in NO production occurred when BAECs were treated with equimolar concentrations of dextran sulfate or heparin, suggesting that the effect of heparin is a consequence of its polyanionic nature rather than its saccharide sequence. The influence of negative charge on NO production was underscored when cells treated with dextran sulfate in the presence of equimolar protamine failed to generate less NO than their respective control. Hirudin, a polyanionic polypeptide, failed to elicit a decrease in NO production by BAECs, suggesting that the sulfate functional groups of the polysaccharides are essential for this effect. Although earlier studies19 20 21 suggested that heparin can alter endothelium-dependent vasoreactivity, this is the first study to show directly that heparin decreases NO production by the endothelial cell and thus leads to an increase in platelet activation.
The results presented here are in no way meant to dispute the well-known beneficial effects of heparin in reducing the thrombotic complications of angioplasty, in reducing reocclusion after thrombolysis, or in reducing graft occlusion after surgical revascularization in the great majority of patients. The potential pathophysiological importance of the ability of heparin to alter NO production by the endothelium is underscored by peripheral-vascular and cardiovascular studies35 36 in which heparin, used as an anticoagulant during procedures, is associated with a number of thrombotic complications that cannot be explained on the basis of technical difficulties or other well-known prothrombotic abnormalities. Donaldson and colleagues35 reported that after infrainguinal bypass surgery, 8% of primary graft failures (7 of 92) could not be attributed to either intrinsic or extrinsic causes. Detailed studies examining the incidence of graft thrombosis after coronary artery bypass surgery have not been performed; however, angiographic evidence of acute thrombosis after coronary angioplasty despite heparin therapy and an acceptable angiographic outcome suggests that prothrombotic mechanisms may account for this adverse action.36 Studies such as these, which attempted to isolate specific abnormalities to explain nontechnical thromboses, document the existence of an enigmatic group of patients who suffer thrombotic events after cardiovascular procedures despite adequate heparin therapy. We therefore postulate that one potential explanation for these thrombotic events may be related to the effects of heparin, a polyanionic glycosaminoglycan, on production by the endothelium of NO, an endothelial product known to inhibit platelet activation and platelet-mediated thrombosis.37
Heparin, composed of alternating units of d-glucosamine and iduronic acid, is a naturally occurring glycosaminoglycan found in mast cells38 that has been shown to manifest a number of other functions independent of its anticoagulant activity. Although endogenously derived heparin is not found in the circulation, heparin-like molecules (heparans) are known to exist on the endothelial cell surface.39 After intravenous administration of heparin, in vivo studies have shown that 88% and 51% of heparin at 2.4 and 6 minutes, respectively, is associated with the endothelium.40
After interacting with the cell surface, heparin has been shown to protect the endothelium from damage by free radicals.41 42 Heparin and other related polyanions, such as dermatan and heparan sulfate, are also capable of releasing diamine oxidase and SOD from the endothelial cell surface.43 44 The release of endothelium-bound SOD, an enzyme that has been shown to increase the biological activity of EDRF/NO,45 46 presumably by preventing inactivation of NO by superoxide, might be important in limiting the ability of the cell to maintain bioactive NO despite the ability of heparin to inactivate free radicals directly. The release of protective SOD from the cell surface might also alter the local environment by allowing free radicals, such as superoxide and hydroxyl radicals, to interact with the cell and in turn influence events such as redox-sensitive gene transcription.
Once bound to the endothelial surface, heparin might also alter the production of NO by the endothelium via a second-messenger mechanism. Studies have shown that heparin, after binding to epithelial cells, leads to a decrease in inositol triphosphate production, which in turn can decrease intracellular calcium concentrations.47 48 Thus, since the activity of eNOS is calcium dependent, the binding of heparin to the cell surface might lead to a decrease in NO synthase activity by an effect on cell signaling.
Although this is the first study to show that heparin alters steady-state Nos 3 mRNA, possibly by modulating transcription, a number of studies have demonstrated that heparin regulates other intracellular events, including gene transcription.49 50 Imai and colleagues,49 using cultured BAECs, have shown that heparin inhibits the expression of the proto-oncogene c-fos. Using Northern blot analysis, they also showed that 20 U/mL heparin reduces steady-state basal as well as thrombin-stimulated pp-ET-1 mRNA expression after 60 minutes. Further studies by Yokokawa and colleagues50 in human umbilical vein endothelial cells confirmed that 10 or 20 U/mL heparin suppresses basal and thrombin-stimulated pp-ET-1 mRNA expression and that this expression is coincident with a rise in endothelin-1 peptide, a potent vasoconstrictor. They postulated that the downregulation of pp-ET-1 transcription was NO dependent, for when they treated cells with L-NMMA, a competitive inhibitor of NO synthase, the decreased expression of pp-ET-1 was abolished. Although both of these studies illustrate the ability of heparin to alter gene transcription, the high doses of heparin used in these experiments make them less biologically relevant than the results reported here.
The present study demonstrates that endothelial cells exposed to heparin produce less NO and that this leads to an increase in platelet aggregation. We suggest that the decrease in NO production by BAECs after exposure to heparin is a result of the combined effects of destabilization of extracellular NO by the release of SOD from the cell surface, by reduced eNOS activity secondary to reduced intracellular calcium, and by reduced steady-state Nos 3 mRNA levels. The nonspecific free radical–scavenging activity of heparin may account for the comparatively modest reduction in S-nitrosothiol concentrations in these experiments in the face of more significant reductions in steady-state Nos 3 mRNA and eNOS protein levels. The relative embarrassment of NO production by BAECs after exposure to heparin over a short period of time might serve as one possible explanation for thrombotic events in cardiovascular settings during which heparin therapy is felt to be adequate. Further clinical studies will be required to address directly the relevance of these observations to thrombotic events in vivo.
Selected Abbreviations and Acronyms
|BAEC||=||bovine aortic endothelial cell|
|ECMB||=||endothelial cell on microcarrier bead|
|eNOS||=||endothelial NO synthase|
|NCS||=||newborn calf serum|
This work was supported in part by National Institutes of Health (NIH) grants HL-53919, HL-48743, and P50-HL-55993; by a Merit Review Award from the US Veterans Administration; and by a grant from Nitro Med, Inc. Dr Upchurch is the recipient of a National Research Service Award from the NIH (HL-09124) and is an American College of Surgeons Resident Scholar. Dr Welch is supported by an NIH Cardiovascular Training Grant Fellowship (P32-HL-07224). Dr Keaney is the recipient of a Clinical Investigator Development Award (HL-03195) from the NIH. We are to grateful to Stephanie Tribuna for her assistance in the preparation of the manuscript.
- Received July 23, 1996.
- Revision received November 19, 1996.
- Accepted November 25, 1996.
- Copyright © 1997 by American Heart Association
Howell WH. Heparin, an anticoagulant: preliminary communication. Am J Physiol. 1922;63:434-435.
Best CH, Cowan C, MacLean DL. Heparin and the formation of white thrombi. J Physiol. 1938;92:20-30.
Eidt JF, Allison P, Noble S, Ashton J, Golino P, McNatt J, Buja LM, Willerson JT. Thrombin is an important mediator of platelet aggregation in stenosed canine coronary arteries with endothelial injury. J Clin Invest. 1989;84:18-27.
Hanson SR, Harker LA. Interruption of acute platelet-dependent thrombosis by the synthetic antithrombin D-phenylalanyl-L-prolyl-L-arginyl chloromethyl ketone. Proc Natl Acad Sci U S A. 1988;85:3184-3188.
Kelly AB, Marzec UM, Krupski W, Bass A, Cadroy Y, Hanson SR, Harker LA. Hirudin interruption of heparin-resistant arterial thrombus formation in baboons. Blood. 1991;77:1006-1012.
Heras M, Chesebro JH, Penny WJ, Bailey KR, Badimon L, Fuster V. Effects of thrombin inhibition on the development of acute platelet-thrombus deposition during angioplasty in pigs: heparin versus recombinant hirudin, a specific thrombin inhibitor. Circulation. 1989;79:657-665.
Heras M, Chesebro JH, Webster MW, Mruk JS, Grill DE, Fuster V. Antithrombotic efficacy of low-molecular-weight heparin in deep arterial injury. Arterioscler Thromb. 1992;12:250-255.
Upchurch GR Jr, Valeri CR, Khuri SF, Rohrer MJ, Welch GN, MacGregor H, Ragno H, Francis S, Rodino LJ, Michelson AD, Loscalzo J. The effect of heparin on fibrinolytic activity and platelet function in vivo. Am J Physiol (Heart Circ Physiol). 1996;271:H528-H534.
Ignarro LJ, Buga GM, Wood KS, Byrns RE, Chaudhuri G. Endothelium-derived relaxing factor produced and released from artery and vein is nitric oxide. Proc Natl Acad Sci U S A. 1987;84:9265-9269.
Light JT Jr, Bellan JA, Roberts MP, Force SD, Chen IL, Kerstein MD, Kadowitz PJ, McNamara DB. Heparin treatment enhancesthe recovery of neoendothelial acetylcholine-induced vascular relaxation after balloon catheter injury in the rabbit aorta. Circulation. 1993;88(suppl II):II-413-II-419.
Jaffe EA, Nachman RL, Becker CG, Minick CR. Culture of human endothelial cells derived from umbilical veins: identification by morphologic and immunologic criteria. J Clin Invest. 1973;52:2745-2756.
Jaffe EA, Hoyer LW, Nachman RL. Synthesis of antihemophilic factor antigen by cultured human endothelial cells. J Clin Invest. 1973;52:2757-2764.
Folkman J, Klagsburn M. Angiogenic factors. Science. 1987;235:442-447.
Davies PF, Truskey GA, Warren HB, O’Connor SE, Eisenhaure BH. Metabolic cooperation between vascular endothelial cells and smooth muscle cells in co-culture: changes in low density lipoprotein metabolism. J Cell Biol. 1985;101:871-879.
Freedman JE, Fabian A, Loscalzo J. Impaired EDRF production by endothelial cells exposed to fibrin monomer and FDP. Am J Physiol. 1995;268:C520-C526.
Stamler J, Mendelsohn ME, Amarante D, Smick D, Andon N, Davies PF, Cooke JP, Loscalzo J. N-Acetylcysteine potentiates platelet inhibition by endothelium-derived relaxing factor. Circ Res. 1989;65:789-795.
Stamler JS, Jaraki O, Osborne J, Simon DI, Keaney J, Vita J, Singel D, Valeri CR, Loscalzo J. Nitric oxide circulates in mammalian plasma primarily as an S-nitroso adduct of serum albumin. Proc Natl Acad Sci U S A. 1992;89:7674-7677.
Lamas S, Marsden PA, Li GK, Tempst P, Michel T. Endothelial cell nitric oxide synthase: molecular cloning and characterization of a distinctive constitutive enzyme isoform. Proc Natl Acad Sci U S A. 1992;89:6348-6352.
Holmes DR, Schwartz RS, Webster MW. Coronary restenosis: what have we learned from angiography? J Am Coll Cardiol. 1991;17:14B-22B.
Mendelsohn ME, O’Neill S, George D, Loscalzo J. Inhibition of fibrinogen binding to human platelets by S-nitroso-N-acetylcysteine. J Biol Chem. 1990;265:19028-19034.
Rosenburg RD, Fritze LMS, Castellot JJ Jr, Karnovsky MJ. Heparin-like molecules as regulators of atherogenesis. Nouv Rev Fr Hematol. 1984;26:255-260.
Hiebert LM, Wice SM, McDuffie NM, Jaques LB. The heparin target organ: the endothelium: studies in a rat model. Q J Med. 1993;86:341-348.
Hiebert LM, Liu J. The protective action of polyelectrolytes on endothelium. Semin Thromb Hemost. 1991;17(suppl 1):42-46.
Robinson-White A, Baylin SB, Olivecrona T, Beaven MA. Binding of diamine oxidase activity to rat and guinea pig microvascular endothelial cells: comparison with lipoprotein lipase binding. J Clin Invest. 1985;76:93-100.
Karlsson K, Marklund SL. Heparin-induced release of extracellular superoxide dismutase to human blood plasma. Biochem J. 1987;242:55-59.
Rubanyi GM, Vanhoutte P. Superoxide anions and hypoxia inactivate endothelium-derived relaxing factor. Am J Physiol. 1986;250:H822-H827.
Boitano S, Dirksen ER, Sanderson MJ. Intercellular propagation of calcium waves mediated by inositol phosphate. Science. 1992;258:292-295.
Dinerman JL, Lowenstein CJ, Snyder SH. Molecular mechanisms of nitric oxide regulation: potential relevance to cardiovascular disease. Circ Res. 1993;73:217-222.
Imai T, Hirata Y, Marumo F. Heparin inhibits endothelin-1 and proto-oncogene c-fos gene expression in cultured bovine endothelial cells. J Cardiovasc Pharmacol. 1993;22:S49-S52.
Yokokawa K, Tahara H, Kohno M, Mandal AK, Yanagisawa M, Takeda T. Heparin regulates endothelin production through endothelium-derived nitric oxide in human endothelial cells. J Clin Invest. 1993;92:2080-2085.